leecb72/LlamaIndex-agentic-workflows

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

Langfuse tracing

The application creates one run-alfred-agent trace for each agent run. The smolagents integration automatically nests model generations and tool calls, including model and token-usage details when the provider returns them.

Configure these environment variables before starting the app:

export LANGFUSE_PUBLIC_KEY="pk-lf-..."
export LANGFUSE_SECRET_KEY="sk-lf-..."
export LANGFUSE_BASE_URL="https://us.cloud.langfuse.com"
export LANGFUSE_TRACING_ENVIRONMENT="development"

Use the base URL for your Langfuse region or self-hosted instance. Keep API keys in your deployment's secret store; do not commit them. If credentials are absent or fail authentication, the app continues to run with tracing disabled and emits a startup warning.

To run sample AlfredAgent in huggingface.co

Git url: https://huggingface.co/spaces/leecb72/AlfredAgent Click App to run // cannot run locally in Cursor Click Files to edit Click three rows for log Click there dots for Settings --> Key and others Variables LANGFUSEPUBLICKEY LANGFUSESECRETKEY LANGFUSEHOST: https://us.cloud.langfuse.com Secrets HFTOKEN from https://huggingface.co/settings/tokens

APPLY LiteLLMModel for free access

To run sample AlfredAgent locally

git clone https://huggingface.co/spaces/leecb72/AlfredAgent cd AlfredAgent

python3 -m venv venv source venv/bin/activate

pip install -r requirements.txt

python app.py